Autogenous bone grafting is widely used to treat fractures, non-unions, and to induce arthrodeses. However, harvest of autogenous bone causes significant morbidity, including prolonged surgery and rehabilitation, scars, blood loss, pain, and infection risk. Bone marrow contains cells which contribute significantly to bone healing in vivo, and to the efficacy of autogenous bone grafts. Under selected in vitro conditions, stromal cells will form colonies which express alkaline phosphatase (CFU-APs). Among these cells are osteoblastic progenitors (CFU-Os), which can be assayed in vitro using markers of bone formation. The applicant believes that aspirated autogenous bone marrow is a valuable source of CFU-Os and that enriching a graft site with CFU-Os will significantly improve the effectiveness of many bone graft substitute materials, including those delivering growth factors. In fact, addition of bone marrow cells may be necessary for optimal performance. Therefore, optimal use of bone marrow may significantly reduce the risk and cost of many grafting procedures. Consequently, this proposal will attempt to answer several important questions: 1) does bone marrow improve the efficacy of available matrix materials? Two canine spinal fusion experiments have been designed to answer this question using the most commonly available allograft matrix materials; 2-3) can bone marrow be further improved by processing in the operating room? Does increasing the number of progenitors transplanted enhance graft performance? Four additional spinal fusion experiments will evaluate methods for rapid intraoperative concentration of CFU-O's from bone marrow; 4) which of the available carrier matrices work best as a delivery system for bone marrow? Five matrices will be evaluated in vitro: coralline hydroxyapatite; mineralized, demineralized, and sintered allograft bone; and guanidine HCL extracted cancellous bone. Matrices will be compared based on their capacity to bind and retain CFU-APs and on their ability to support the proliferation and differentiation of CFU-Os. In vitro performance will be correlated with in vivo performance in the canine spinal fusion model; 5) what clinical factors influence the number and function of CFU-Os? Marrow will be harvested from 200 patients over four years. The number and prevalence of CFU-APs will be assayed. In addition, long-term cultures from each patient will be assayed to quantify proliferation of CFU-APs and osteoblastic differentiation based on osteocalcin and bone sialoprotein synthesis and mineralization. Using this data, the influence of age, sex, and selected disease states on the number and function of osteoblastic progenitors will be assessed; and 6) what is the prevalence of human CFU-Os among CFU-APs? Marrow samples from each patient and canine subject will be assayed to quantify the prevalence of CFU-Os present.